The rate of anal carcinoma among HIV positive men is 70/100,000 person years. This is higher than the rate of cervical carcinoma prior to the onset of routine cytology screening. Anal squamous intra-epithelial lesions (ASIL) are pre-malignant lesions that can be found by screening cytology analogous to cervical dysplasia. It is felt that persistent infection with oncogenic subtypes of human papillomavirus (HPV) is the most important cofactor for development of anal carcinoma. Prior to the routine use of highly active anti-retroviral therapy, screening for anal dysplasia was studied in large cohorts of predominantly white HIV+ and HIV- men who have sex with men. High rates of ASIL and anal infection with HPV were found. It is not clear how this applies to other groups of HIV+ men including minorities, heterosexual men, and intravenous drug users, and how the rate of HPV and anal dysplasia may have changed after widespread use of HAART. This study will enroll 350 HIV+ men and 200 HIV- men who are 40% Hispanic and 40% African-American. Forty percent of the HIV+ men will report intravenous drug use or sex with women as their risk behavior for HIV. A standardized questionnaire will be administered to determine past sexual behaviors. Each participant will undergo anal cytology and anal HPV testing. Those individuals with oncogenic HPV at the first test will be followed with anal cytology and HPV testing every six months for a total of two years or five exams. Each participant with abnormal cytology will undergo high-resolution anoscopy with biopsy. Factors associated with prevalent ASIL and HPV infection including immunologic and sexual parameters will be compared using logistic regression. Persistence of HPV will be defined as the time to the first negative test for HPV and will be analyzed using Cox proportional hazards model. Factors associated with persistence will be included in a multiple Cox model. The influence of HAART will be assessed in both the prevalence and prospective components by including a term in the multiple models for length of time on anti-retroviral therapy with suppression of HIV. This study will help provide important new information on which groups of HIV+ men may benefit from screening and how HAART may modify this risk.